I am looking at yearly data sets to quantify the impact of abiotic factors on honey bee flight. It is obvious that at certain times of the year (winter) the temperature is the most important where as it is solar radiation and light intensity (summer) for other times. I could arbitrarily divide the year into four parts based on calendar seasons, but this would not accurately reflect the abiotic conditions. Is there a way to determine from what value an abiotic factor becomes the dominant influence? I do not have a strong statistics background so I am hoping for some guidance. Thank you.
Dear Alexandra,
Simple linear regression is the least squares estimator of a linear regression model with a single explanatory variable. The alternative terms explanatory variable, independent variable, or predictor, are used in a regression analysis. As far as I can understand there are one dependent variable (honey bee flight) and three independent variables (temperature, radiation and light intensity). In this case, the regression will not be simple. The regression will be simple for single independent variable (for example, temperature) only. To include the seasonal variation on independent variable in simple regression it's necessary to combine some simple regressions for certain times of the year by weight factors. But it is better in my opinion to use a single regressions for each season separately.
Attach for me the data file and I'll try to help you in building the regression.
Kind regards
Alexandra,
I guess that what you need is some kind of time series analysis, difficult to say more without the data.
Cheers,
Zivan
Hello,
Look up about detrending time series. Assuming monthly data, the simplest way is applying this multiple regression
yi = w0 + w1 ti + w2 sin(2πti / 12) + w3 cos(2πti / 12) + εi
but, as Zivan said, it is difficult to say which would be the best model
Bye
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Environmental Science